Reviewed-by: Tom Cosgrove <tom.cosgrove@arm.com>
Reviewed-by: Richard Levitte <levitte@openssl.org>
Reviewed-by: Matt Caswell <matt@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/20519)
(cherry picked from commit 110dac5783)
RISC-V currently only offers a GMULT() callback for accelerated
processing. Let's implement the missing piece to have GHASH()
available as well. Like GMULT(), we provide a variant for
systems with the Zbkb extension (including brev8).
The integration follows the existing pattern for GMULT()
in RISC-V. We keep the C implementation as we need to decide
if we can call an optimized routine at run-time.
The C implementation is the fall-back in case we don't have
any extensions available that can be used to accelerate
the calculation.
Tested with all combinations of possible extensions
on QEMU (limiting the available instructions accordingly).
No regressions observed.
Signed-off-by: Christoph Müllner <christoph.muellner@vrull.eu>
Reviewed-by: Tomas Mraz <tomas@openssl.org>
Reviewed-by: Paul Dale <pauli@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/20078)
The existing GCM calculation provides some potential
for further optimizations. Let's use the demo code
from the RISC-V cryptography extension groups
(https://github.com/riscv/riscv-crypto), which represents
the extension architect's intended use of the clmul instruction.
The GCM calculation depends on bit and byte reversal.
Therefore, we use the corresponding instructions to do that
(if available at run-time).
The resulting computation becomes quite compact and passes
all tests.
Note, that a side-effect of this change is a reduced register
usage in .gmult(), which opens the door for an efficient .ghash()
implementation.
Signed-off-by: Christoph Müllner <christoph.muellner@vrull.eu>
Reviewed-by: Tomas Mraz <tomas@openssl.org>
Reviewed-by: Paul Dale <pauli@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/20078)
A recent commit introduced a Perl module for common code.
This patch changes the GCM code to use this module, removes duplicated code,
and moves the instruction encoding functions into the module.
Signed-off-by: Christoph Müllner <christoph.muellner@vrull.eu>
Reviewed-by: Tomas Mraz <tomas@openssl.org>
Reviewed-by: Paul Dale <pauli@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/20078)
In RISC-V we have multiple extensions, that can be
used to accelerate processing.
The known extensions are defined in riscv_arch.def.
From that file test functions of the following
form are generated: RISCV_HAS_$ext().
In recent commits new ways to define the availability
of these test macros have been defined. E.g.:
#define RV32I_ZKND_ZKNE_CAPABLE \
(RISCV_HAS_ZKND() && RISCV_HAS_ZKNE())
[...]
#define RV64I_ZKND_ZKNE_CAPABLE \
(RISCV_HAS_ZKND() && RISCV_HAS_ZKNE())
This leaves us with two different APIs to test capabilities.
Further, creating the same macros for RV32 and RV64 results
in duplicated code (see example above).
This inconsistent situation makes it hard to integrate
further code. So let's clean this up with the following steps:
* Replace RV32I_* and RV64I_* macros by RICSV_HAS_* macros
* Move all test macros into riscv_arch.h
* Use "AND" and "OR" to combine tests with more than one extension
* Rename include files for accelerated processing (remove extension
postfix).
We end up with compile time tests for RV32/RV64 and run-time tests
for available extensions. Adding new routines (e.g. for vector crypto
instructions) should be straightforward.
Testing showed no regressions.
Signed-off-by: Christoph Müllner <christoph.muellner@vrull.eu>
Reviewed-by: Tomas Mraz <tomas@openssl.org>
Reviewed-by: Paul Dale <pauli@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/20078)
These aren't currently checked when they are called in cipher_aes_gcm_hw_armv8.inc,
but they are declared as returning as size_t the number of bytes they have processed,
and the aes_gcm_*_*_kernel (unroll by 4) versions of these do return the correct
values.
Change-Id: Ic3eaf139e36e29e8779b5bd8b867c08fde37a337
Reviewed-by: Paul Dale <pauli@openssl.org>
Reviewed-by: Tomas Mraz <tomas@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/20191)
Signed-off-by: Xu Yizhou <xuyizhou1@huawei.com>
Reviewed-by: Hugo Landau <hlandau@openssl.org>
Reviewed-by: Tomas Mraz <tomas@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/19619)
Since OPENSSL_malloc() and friends report ERR_R_MALLOC_FAILURE, and
at least handle the file name and line number they are called from,
there's no need to report ERR_R_MALLOC_FAILURE where they are called
directly, or when SSLfatal() and RLAYERfatal() is used, the reason
`ERR_R_MALLOC_FAILURE` is changed to `ERR_R_CRYPTO_LIB`.
There were a number of places where `ERR_R_MALLOC_FAILURE` was reported
even though it was a function from a different sub-system that was
called. Those places are changed to report ERR_R_{lib}_LIB, where
{lib} is the name of that sub-system.
Some of them are tricky to get right, as we have a lot of functions
that belong in the ASN1 sub-system, and all the `sk_` calls or from
the CRYPTO sub-system.
Some extra adaptation was necessary where there were custom OPENSSL_malloc()
wrappers, and some bugs are fixed alongside these changes.
Reviewed-by: Tomas Mraz <tomas@openssl.org>
Reviewed-by: Hugo Landau <hlandau@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/19301)
s390x GHASH assembler implementation assumed it was called from a
gcm128_context structure where the Xi paramter to the ghash function was
embedded in that structure. Since the structure layout resembles the paramter
block required for kimd-GHASH, the assembler code simply assumed the 128 bytes
after Xi are the hash subkey.
This assumption was broken with the introduction of AES-GCM-SIV which uses the
GHASH implementation without a gcm128_context structure. Furthermore, the
bytes following the Xi input parameter to the GHASH function do not contain
the hash subkey. To fix this, we remove the assumption about the calling
context and build the parameter block on the stack. This requires some
copying of data to and from the stack. While this introduces a performance
degradation, new systems anyway use kma for GHASH/AES-GCM.
Finally fixes#18693 for s390x.
Signed-off-by: Juergen Christ <jchrist@linux.ibm.com>
Reviewed-by: Todd Short <todd.short@me.com>
Reviewed-by: Matt Caswell <matt@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/18939)
Fixes#16721
This uses AES-ECB to create a counter mode AES-CTR32 (32bit counter, I could
not get AES-CTR to work as-is), and GHASH to implement POLYVAL. Optimally,
there would be separate polyval assembly implementation(s), but the only one
I could find (and it was SSE2 x86_64 code) was not Apache 2.0 licensed.
This implementation lives only in the default provider; there is no legacy
implementation.
The code offered in #16721 is not used; that implementation sits on top of
OpenSSL, this one is embedded inside OpenSSL.
Full test vectors from RFC8452 are included, except the 0 length plaintext;
that is not supported; and I'm not sure it's worthwhile to do so.
Reviewed-by: Hugo Landau <hlandau@openssl.org>
Reviewed-by: Tomas Mraz <tomas@openssl.org>
Reviewed-by: Paul Dale <pauli@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/18693)
Rework of GCM code did not include s390x causing NULL pointer dereferences on
GCM operations other than AES-GCM on platforms that support kma. Fix this by
a proper setup of the function pointers.
Fixes: 92c9086e5c ("Use separate function to get GCM functions")
Signed-off-by: Juergen Christ <jchrist@linux.ibm.com>
Reviewed-by: Paul Dale <pauli@openssl.org>
Reviewed-by: Hugo Landau <hlandau@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/18862)
Fixes openssl#18073.
Reviewed-by: Tomas Mraz <tomas@openssl.org>
Reviewed-by: Todd Short <todd.short@me.com>
Reviewed-by: Paul Dale <pauli@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/18327)
Rename x86-32 assembly files from .s to .S. While processing the .S file
gcc will use the pre-processor whic will evaluate macros and ifdef. This
is turn will be used to enable the endbr32 opcode based on the __CET__
define.
Signed-off-by: Sebastian Andrzej Siewior <sebastian@breakpoint.cc>
Reviewed-by: Tomas Mraz <tomas@openssl.org>
Reviewed-by: Paul Dale <pauli@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/18353)
ghash-riscv64.pl implements 128-bit galois field multiplication for
use in the GCM mode using RISC-V carryless multiplication primitives.
The clmul-accelerated routine can be selected by setting the Zbb and
Zbc bits of the OPENSSL_riscvcap environment variable at runtime.
Reviewed-by: Philipp Tomsich <philipp.tomsich@vrull.eu>
Signed-off-by: Henry Brausen <henry.brausen@vrull.eu>
Reviewed-by: Tomas Mraz <tomas@openssl.org>
Reviewed-by: Paul Dale <pauli@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/17640)
Decoration prefix for some assembler labels in aes-gcm-avx512.pl was
fixed for mingw64 build.
Reviewed-by: Matt Caswell <matt@openssl.org>
Reviewed-by: Tomas Mraz <tomas@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/17868)
The EOR3 instruction is implemented with .inst, and the code here is enabled
using run-time detection of the CPU capabilities, so no need to explicitly
ask for the sha3 extension.
Fixes#17773
Reviewed-by: Kurt Roeckx <kurt@roeckx.be>
Reviewed-by: Tomas Mraz <tomas@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/17795)
Increase the block numbers to 8 for every iteration. Increase the hash
table capacity. Make use of EOR3 instruction to improve the performance.
This can improve performance 25-40% on out-of-order microarchitectures
with a large number of fast execution units, such as Neoverse V1. We also
see 20-30% performance improvements on other architectures such as the M1.
Assembly code reviewd by Tom Cosgrove (ARM).
Reviewed-by: Bernd Edlinger <bernd.edlinger@hotmail.de>
Reviewed-by: Paul Dale <pauli@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/15916)
Assembly code reviewed by Shricharan Srivatsan <ssrivat@us.ibm.com>
Reviewed-by: Tomas Mraz <tomas@openssl.org>
Reviewed-by: Paul Dale <pauli@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/16854)
GCC's __ARMEL__ and __ARMEB__ defines denote little- and big-endian arm,
respectively. They are not defined on aarch64, which instead use
__AARCH64EL__ and __AARCH64EB__.
However, OpenSSL's assembly originally used the 32-bit defines on both
platforms and even define __ARMEL__ and __ARMEB__ in arm_arch.h. This is
less portable and can even interfere with other headers, which use
__ARMEL__ to detect little-endian arm.
Over time, the aarch64 assembly has switched to the correct defines,
such as in 32bbb62ea6. This commit
finishes the job: poly1305-armv8.pl needed a fix and the dual-arch
armx.pl files get one more transform to convert from 32-bit to 64-bit.
(There is an even more official endianness detector, __ARM_BIG_ENDIAN in
the Arm C Language Extensions. But I've stuck with the GCC ones here as
that would be a larger change.)
Reviewed-by: Matt Caswell <matt@openssl.org>
Reviewed-by: Tomas Mraz <tomas@openssl.org>
Reviewed-by: Paul Dale <pauli@openssl.org>
Reviewed-by: Bernd Edlinger <bernd.edlinger@hotmail.de>
(Merged from https://github.com/openssl/openssl/pull/17373)
This change adds optional support for
- Armv8.3-A Pointer Authentication (PAuth) and
- Armv8.5-A Branch Target Identification (BTI)
features to the perl scripts.
Both features can be enabled with additional compiler flags.
Unless any of these are enabled explicitly there is no code change at
all.
The extensions are briefly described below. Please read the appropriate
chapters of the Arm Architecture Reference Manual for the complete
specification.
Scope
-----
This change only affects generated assembly code.
Armv8.3-A Pointer Authentication
--------------------------------
Pointer Authentication extension supports the authentication of the
contents of registers before they are used for indirect branching
or load.
PAuth provides a probabilistic method to detect corruption of register
values. PAuth signing instructions generate a Pointer Authentication
Code (PAC) based on the value of a register, a seed and a key.
The generated PAC is inserted into the original value in the register.
A PAuth authentication instruction recomputes the PAC, and if it matches
the PAC in the register, restores its original value. In case of a
mismatch, an architecturally unmapped address is generated instead.
With PAuth, mitigation against ROP (Return-oriented Programming) attacks
can be implemented. This is achieved by signing the contents of the
link-register (LR) before it is pushed to stack. Once LR is popped,
it is authenticated. This way a stack corruption which overwrites the
LR on the stack is detectable.
The PAuth extension adds several new instructions, some of which are not
recognized by older hardware. To support a single codebase for both pre
Armv8.3-A targets and newer ones, only NOP-space instructions are added
by this patch. These instructions are treated as NOPs on hardware
which does not support Armv8.3-A. Furthermore, this patch only considers
cases where LR is saved to the stack and then restored before branching
to its content. There are cases in the code where LR is pushed to stack
but it is not used later. We do not address these cases as they are not
affected by PAuth.
There are two keys available to sign an instruction address: A and B.
PACIASP and PACIBSP only differ in the used keys: A and B, respectively.
The keys are typically managed by the operating system.
To enable generating code for PAuth compile with
-mbranch-protection=<mode>:
- standard or pac-ret: add PACIASP and AUTIASP, also enables BTI
(read below)
- pac-ret+b-key: add PACIBSP and AUTIBSP
Armv8.5-A Branch Target Identification
--------------------------------------
Branch Target Identification features some new instructions which
protect the execution of instructions on guarded pages which are not
intended branch targets.
If Armv8.5-A is supported by the hardware, execution of an instruction
changes the value of PSTATE.BTYPE field. If an indirect branch
lands on a guarded page the target instruction must be one of the
BTI <jc> flavors, or in case of a direct call or jump it can be any
other instruction. If the target instruction is not compatible with the
value of PSTATE.BTYPE a Branch Target Exception is generated.
In short, indirect jumps are compatible with BTI <j> and <jc> while
indirect calls are compatible with BTI <c> and <jc>. Please refer to the
specification for the details.
Armv8.3-A PACIASP and PACIBSP are implicit branch target
identification instructions which are equivalent with BTI c or BTI jc
depending on system register configuration.
BTI is used to mitigate JOP (Jump-oriented Programming) attacks by
limiting the set of instructions which can be jumped to.
BTI requires active linker support to mark the pages with BTI-enabled
code as guarded. For ELF64 files BTI compatibility is recorded in the
.note.gnu.property section. For a shared object or static binary it is
required that all linked units support BTI. This means that even a
single assembly file without the required note section turns-off BTI
for the whole binary or shared object.
The new BTI instructions are treated as NOPs on hardware which does
not support Armv8.5-A or on pages which are not guarded.
To insert this new and optional instruction compile with
-mbranch-protection=standard (also enables PAuth) or +bti.
When targeting a guarded page from a non-guarded page, weaker
compatibility restrictions apply to maintain compatibility between
legacy and new code. For detailed rules please refer to the Arm ARM.
Compiler support
----------------
Compiler support requires understanding '-mbranch-protection=<mode>'
and emitting the appropriate feature macros (__ARM_FEATURE_BTI_DEFAULT
and __ARM_FEATURE_PAC_DEFAULT). The current state is the following:
-------------------------------------------------------
| Compiler | -mbranch-protection | Feature macros |
+----------+---------------------+--------------------+
| clang | 9.0.0 | 11.0.0 |
+----------+---------------------+--------------------+
| gcc | 9 | expected in 10.1+ |
-------------------------------------------------------
Available Platforms
------------------
Arm Fast Model and QEMU support both extensions.
https://developer.arm.com/tools-and-software/simulation-models/fast-modelshttps://www.qemu.org/
Implementation Notes
--------------------
This change adds BTI landing pads even to assembly functions which are
likely to be directly called only. In these cases, landing pads might
be superfluous depending on what code the linker generates.
Code size and performance impact for these cases would be negligible.
Interaction with C code
-----------------------
Pointer Authentication is a per-frame protection while Branch Target
Identification can be turned on and off only for all code pages of a
whole shared object or static binary. Because of these properties if
C/C++ code is compiled without any of the above features but assembly
files support any of them unconditionally there is no incompatibility
between the two.
Useful Links
------------
To fully understand the details of both PAuth and BTI it is advised to
read the related chapters of the Arm Architecture Reference Manual
(Arm ARM):
https://developer.arm.com/documentation/ddi0487/latest/
Additional materials:
"Providing protection for complex software"
https://developer.arm.com/architectures/learn-the-architecture/providing-protection-for-complex-software
Arm Compiler Reference Guide Version 6.14: -mbranch-protection
https://developer.arm.com/documentation/101754/0614/armclang-Reference/armclang-Command-line-Options/-mbranch-protection?lang=en
Arm C Language Extensions (ACLE)
https://developer.arm.com/docs/101028/latest
Addional Notes
--------------
This patch is a copy of the work done by Tamas Petz in boringssl. It
contains the changes from the following commits:
aarch64: support BTI and pointer authentication in assembly
Change-Id: I4335f92e2ccc8e209c7d68a0a79f1acdf3aeb791
URL: https://boringssl-review.googlesource.com/c/boringssl/+/42084
aarch64: Improve conditional compilation
Change-Id: I14902a64e5f403c2b6a117bc9f5fb1a4f4611ebf
URL: https://boringssl-review.googlesource.com/c/boringssl/+/43524
aarch64: Fix name of gnu property note section
Change-Id: I6c432d1c852129e9c273f6469a8b60e3983671ec
URL: https://boringssl-review.googlesource.com/c/boringssl/+/44024
Change-Id: I2d95ebc5e4aeb5610d3b226f9754ee80cf74a9af
Reviewed-by: Paul Dale <pauli@openssl.org>
Reviewed-by: Tomas Mraz <tomas@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/16674)
For functions that exist in 1.1.1 provide a simple aliases via #define.
Fixes#15236
Functions with OSSL_DECODER_, OSSL_ENCODER_, OSSL_STORE_LOADER_,
EVP_KEYEXCH_, EVP_KEM_, EVP_ASYM_CIPHER_, EVP_SIGNATURE_,
EVP_KEYMGMT_, EVP_RAND_, EVP_MAC_, EVP_KDF_, EVP_PKEY_,
EVP_MD_, and EVP_CIPHER_ prefixes are renamed.
Reviewed-by: Paul Dale <pauli@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/15405)
libimplementations.a was a nice idea, but had a few flaws:
1. The idea to have common code in libimplementations.a and FIPS
sensitive helper functions in libfips.a / libnonfips.a didn't
catch on, and we saw full implementation ending up in them instead
and not appearing in libimplementations.a at all.
2. Because more or less ALL algorithm implementations were included
in libimplementations.a (the idea being that the appropriate
objects from it would be selected automatically by the linker when
building the shared libraries), it's very hard to find only the
implementation source that should go into the FIPS module, with
the result that the FIPS checksum mechanism include source files
that it shouldn't
To mitigate, we drop libimplementations.a, but retain the idea of
collecting implementations in static libraries. With that, we not
have:
libfips.a
Includes all implementations that should become part of the FIPS
provider.
liblegacy.a
Includes all implementations that should become part of the legacy
provider.
libdefault.a
Includes all implementations that should become part of the
default and base providers.
With this, libnonfips.a becomes irrelevant and is dropped.
libcommon.a is retained to include common provider code that can be
used uniformly by all providers.
Fixes#15157
Reviewed-by: Tomas Mraz <tomas@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/15171)
The incorrect code is in #ifdef branch that is normally
not compiled in.
Signed-off-by: Zhang Jinde <zjd5536@163.com>
Reviewed-by: Paul Dale <pauli@openssl.org>
Reviewed-by: Tomas Mraz <tomas@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/12968)
This includes error reporting for libcrypto sub-libraries in surprising
places.
This was done using util/err-to-raise
Reviewed-by: Paul Dale <paul.dale@oracle.com>
(Merged from https://github.com/openssl/openssl/pull/13318)
Many of the new types introduced by OpenSSL 3.0 have an OSSL_ prefix,
e.g., OSSL_CALLBACK, OSSL_PARAM, OSSL_ALGORITHM, OSSL_SERIALIZER.
The OPENSSL_CTX type stands out a little by using a different prefix.
For consistency reasons, this type is renamed to OSSL_LIB_CTX.
Reviewed-by: Paul Dale <paul.dale@oracle.com>
Reviewed-by: Matt Caswell <matt@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/12621)
